Cold storage control system and method based on real-time electricity price

Abstract

This invention discloses a cold energy storage control system and method based on real-time electricity prices, belonging to the field of cold energy management technology for smart grids and energy storage systems. It aims to solve problems such as low matching degree between electricity prices and cold energy storage strategies, imbalance between cold energy supply and demand, and lack of dynamic adjustment mechanisms. This invention collects and periodically stores real-time electricity prices from the power grid, extracts features such as historical periodic electricity price peaks, valleys, and peak-valley transition times, and filters historical matching periods to generate electricity price predictions and fluctuation ranges. It also collects operational data from energy storage battery packs, and generates heat dissipation demand prediction results based on historical fitted period data combined with charging and discharging plans and ambient temperature predictions. Furthermore, it divides electricity price peak-valley cold energy storage intervals, constructs temporary cold energy storage intervals as needed, and dynamically adjusts the cold energy storage capacity, achieving precise matching between cold energy storage strategies and electricity prices and heat dissipation demands, thereby improving peak-valley arbitrage profits and system operational economy.

Description

Technical Field

[0001] This invention relates to the field of cooling capacity management technology for smart grids and energy storage systems, and more specifically to a cooling capacity storage control system and method based on real-time electricity prices. Background Technology

[0002] Against the backdrop of the ongoing construction of smart grids and the widespread application of peak-valley electricity pricing mechanisms, energy storage systems, as core facilities for grid peak shaving and valley filling to ensure stable energy supply, are receiving increasing attention for their operational stability and economic efficiency. Energy storage battery packs continuously generate heat during charging and discharging; if heat dissipation is not timely, it will directly affect the battery's lifespan and operational safety. Therefore, cold energy storage has become a key technological means for supporting energy storage systems.

[0003] Currently, the cold storage strategies of energy storage systems largely rely on empirical settings, which have revealed numerous problems in practical applications. In terms of electricity price response, traditional cold storage strategies often make simple decisions based solely on real-time electricity prices, failing to fully consider the cyclical changes and fluctuations in electricity prices. This results in a low match between cold storage timing and electricity price off-peak periods. It may involve cold storage operations during peak electricity price periods, significantly increasing operating costs, or, due to misjudgments of electricity price trends, insufficient cold storage capacity during off-peak periods, failing to meet subsequent heat dissipation needs. Regarding heat dissipation demand adaptation, the heat dissipation demand of energy storage battery packs is significantly affected by factors such as charging and discharging power, operating time, and ambient temperature, exhibiting significant dynamic changes. Traditional methods generally use fixed heat dissipation coefficients to calculate demand, making it difficult to accurately match actual heat dissipation needs under different operating conditions. This leads to an imbalance between cold supply and demand; excess cold storage results in energy waste and equipment redundancy, while insufficient cold storage fails to effectively ensure battery safety. Therefore, to overcome these limitations, this invention proposes a cold storage control system and method based on real-time electricity prices. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the present invention aims to provide a cold energy storage control system and method based on real-time electricity prices. This system solves the problem of how to accurately adapt cold energy storage strategies to changes in electricity prices and the heat dissipation requirements of energy storage battery packs, avoids the risks of electricity price prediction, and resolves the problems of cold energy supply and demand imbalance and lack of dynamic adjustment.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A cold storage control system based on real-time electricity prices includes an electricity price cycle prediction module, a demand prediction module, a cold storage zone division module, and a monitoring and adjustment module.

[0007] The electricity price cycle prediction module is used to acquire real-time electricity price data from the power grid and store it periodically, extract historical cycle electricity price change characteristics, filter historical matching cycles, and combine the current cycle's collected electricity price data change trend with the weighted electricity price data of historical matching cycles to generate the current cycle's predicted electricity price data and the predicted electricity price fluctuation range.

[0008] The demand forecasting module is used to collect the operating data of the energy storage battery pack and, based on the historical operating data of the historical fitting cycle, combined with the charging and discharging plan parameters of the pack in the next cycle, to predict the total heat dissipation demand of the energy storage battery pack in the next cycle and the heat dissipation demand forecast threshold.

[0009] The cold storage zone division module is used to calculate the appropriate cold storage capacity and corresponding appropriate cold storage duration based on the cold storage efficiency of the refrigerant and the rated cooling power. It combines the current cycle's predicted electricity price data and the predicted fluctuation range of electricity price to divide the effective candidate cold storage zones. Based on the benchmark cold storage duration, it corrects the effective candidate cold storage zones and divides the current cycle's peak-valley cold storage zones.

[0010] The monitoring and adjustment module determines whether there is a cold storage gap by predicting the range of the peak-valley cold storage interval for the next electricity price cycle and combining it with real-time monitoring of the cold storage status of the refrigerant. If so, a temporary cold storage interval is constructed. At the same time, based on the current cycle's cold storage margin deviation, a correction value for the cold storage capacity in the next cycle is generated.

[0011] Specifically, the steps for filtering historical matching periods include:

[0012] Acquire real-time electricity price data from the power grid, mark the collection timestamp, periodically store the electricity price data according to a preset constant period, construct an electricity price database, and divide the data storage partition for the current period into a data storage partition for the historical period.

[0013] Historical electricity price data is retrieved from the historical periodic data storage partition of the electricity price database. The historical electricity price data is then filtered based on the periodic time attribute and the working day attribute to obtain historical benchmark data. Electricity price change characteristics are extracted to construct a historical electricity price change feature library of the same period.

[0014] Retrieve the collected electricity price data from the current period data storage partition of the electricity price database, and extract the electricity price change characteristics of the collected period in the current period;

[0015] Calculate the similarity coefficient between the electricity price change characteristics of the current period and the electricity price change characteristics of each historical period; select historical periods with similarity coefficients greater than the preset similarity threshold as historical matching periods, and calculate the average of the similarity coefficients of historical matching periods as an indicator of the fit between the current period's electricity price change trend and the historical electricity price change trend of the same type of period.

[0016] Specifically, the steps for generating the current period's predicted electricity price data and the predicted fluctuation range for electricity prices include:

[0017] The electricity price data of the current period, the electricity price data of the corresponding period of each historical matching period, and the matching degree index are linked and integrated to form a prediction input dataset.

[0018] The historical matching period electricity price data in the prediction input dataset are weighted, and the weighting coefficient is the ratio of the similarity coefficient to the fit index of each historical matching period, to obtain the weighted historical electricity price data.

[0019] Based on the electricity price data collected for the current period, combined with weighted historical electricity price data, an initial electricity price prediction sequence for the remaining time periods of the current period is generated using a time-series data extrapolation method.

[0020] Calculate the deviation correction coefficient between the actual electricity price data of the corresponding prediction period in the historical matching cycle and the predicted electricity price data of the initial electricity price prediction sequence corresponding to the historical matching cycle; calculate the deviation correction coefficient of each historical matching cycle by weighting the similarity coefficient to obtain the comprehensive deviation correction coefficient; correct the deviation of the initial electricity price prediction sequence to form the electricity price prediction sequence for the remaining time of the current cycle.

[0021] Retrieve the predicted electricity price data and the corresponding actual electricity price data for all historical matching periods, and calculate the absolute deviation between the predicted electricity price data and the corresponding actual electricity price data for each historical matching period. By the ratio of the absolute deviation to the actual electricity price for the corresponding period, the relative deviation rate for each period of each historical matching period is obtained.

[0022] The maximum and average relative deviation rates for each time period in all historical matching cycles are statistically analyzed. The maximum value is used as the upper limit parameter of the deviation rate, and the average value is used as the benchmark parameter of the deviation rate.

[0023] Each predicted electricity price in the electricity price prediction sequence is used as a benchmark electricity value; the basic fluctuation threshold corresponding to each benchmark electricity value is calculated based on the deviation rate benchmark parameter, and the basic fluctuation threshold is verified by combining the deviation rate upper limit parameter to obtain the target fluctuation threshold, which is used to define the electricity price prediction fluctuation range at each prediction time.

[0024] Specifically, the steps for predicting the total heat dissipation demand of the energy storage battery pack for the next cycle and the predicted threshold for heat dissipation demand include:

[0025] Real-time acquisition of operational data from energy storage battery packs, marking the acquisition timestamp; and partitioning the operational data into current cycle operational data storage partitions and historical cycle operational data storage partitions;

[0026] Configure the number of fitting cycles, which is used to retrieve historical running data of historical fitting cycles that are a number of fitting cycles away from the current period from the historical running data storage partition of the heat dissipation demand database;

[0027] The retrieved historical operating data are classified according to the charging and discharging power range and the ambient temperature range. For the historical operating data in each category range, the correlation coefficient between the heat dissipation requirement of the charging box and the charging and discharging power and the ambient temperature is calculated, and the correlation curve of the heat dissipation requirement of the charging box with the charging and discharging power and the ambient temperature under different operating conditions is generated.

[0028] Retrieve the running data of the current cycle's plug-in box during the running period, obtain the charging and discharging plan parameters of the plug-in box for the next cycle, and form a set of predictive input parameters. The charging and discharging plan parameters include the set values ​​of charging and discharging power for each time period and ambient temperature data.

[0029] Substitute the charging and discharging power setpoints and ambient temperature data for each period of the next cycle from the predicted input parameter set into the fitted correlation curve to calculate the heat dissipation demand of the charging box for each period of the next cycle; sum them up to obtain the predicted value of the total heat dissipation demand of the charging box in the next cycle.

[0030] Retrieve the predicted and actual values ​​of heat dissipation demand under the same operating conditions in the historical fitting period, and calculate the deviation rate between the predicted and actual values ​​of heat dissipation demand for each period; count the deviation rates of all historical fitting periods, calculate the average and maximum values ​​of the deviation rates, and then calculate the upper and lower boundaries of the heat dissipation demand prediction threshold.

[0031] Specifically, the steps for delineating valid candidate cold storage areas include:

[0032] The system acquires real-time electricity price data and current cycle predicted electricity price data, generates the current cycle electricity price sequence, and calculates the average deviation rate between the predicted and actual electricity price values ​​for the current cycle's collected periods. If the average deviation rate is lower than a preset stability threshold, the current cycle electricity price sequence is determined to be stable.

[0033] Set a trigger time for the cold storage zone. If the trigger time is reached or the current electricity price sequence is determined to be stable, then initiate the electricity price peak-valley cold storage zone division operation:

[0034] Based on the total heat dissipation demand of the next cycle's insertion box and the preset refrigerant storage efficiency, the baseline storage capacity is calculated; based on the rated cooling power and the baseline storage capacity, the baseline storage duration is calculated.

[0035] Based on the upper limit of the predicted heat dissipation demand threshold, the appropriate cold storage capacity is calculated, and combined with the rated cooling power, the appropriate cold storage duration is obtained, forming a reasonable range for the cold storage duration.

[0036] Receive the current cycle's predicted electricity price data and the predicted electricity price fluctuation range, and set a preset valley price threshold; iterate through each time period of the remaining time in the current cycle, and filter out the time periods where the lower limit of the predicted electricity price fluctuation range is lower than the preset valley price threshold, as the basic candidate time periods;

[0037] All basic candidate time periods are sorted in chronological order and integrated to form effective candidate cold storage intervals; the duration of the effective candidate cold storage intervals is then calculated and labeled with duration information.

[0038] Specifically, the steps for dividing the current electricity price peak-valley cold storage range include:

[0039] The duration of each valid candidate cold storage interval is compared with the reasonable range of cold storage duration. If the duration of a valid candidate cold storage interval is within the reasonable range of cold storage duration, the valid candidate cold storage interval with the smallest deviation from the benchmark cold storage duration is selected as the peak-valley cold storage interval for the current cycle.

[0040] If the duration of a valid candidate cold storage interval exceeds the appropriate cold storage duration, the valid candidate cold storage interval will be truncated, and the period starting from the lowest point of the lower limit of the electricity price prediction fluctuation range and having a duration equal to the appropriate cold storage duration will be retained as the electricity price peak-valley cold storage interval.

[0041] If the duration of all valid candidate cold storage intervals in the current cycle does not reach the benchmark cold storage duration, then all valid candidate periods are sorted from low to high according to the lower limit of the electricity price prediction fluctuation range; the duration of each valid candidate period is accumulated until the accumulated duration reaches the benchmark cold storage duration, and the selected valid candidate periods are integrated into the electricity price peak-valley cold storage interval.

[0042] Specifically, the steps for predicting the range of peak-valley electricity prices and cold storage for the next electricity cycle include:

[0043] Based on the periodic time attributes, working day attributes, and electricity price change characteristics of the next cycle, historical matching cycles for the next cycle are selected.

[0044] Retrieve historical data on peak-valley electricity price and cold storage interval schemes, including start time, end time, interval duration, and appropriate cold storage capacity.

[0045] Clustering operations are performed on the start and end times of the electricity price peak-valley cold storage interval for all historical matching cycles to generate corresponding start time clusters and end time clusters. The data centrality within each cluster is calculated, and the start time cluster and end time cluster with the highest data centrality within each cluster are selected as the core start time cluster and core end time cluster.

[0046] Based on the time distribution of the core start time cluster and the core end time cluster, the start time range and end time range of the electricity price peak-valley cold storage interval for the next cycle are generated.

[0047] Specifically, the steps for constructing a temporary cold storage area include:

[0048] The system continuously collects real-time refrigerant storage capacity and marks the collection timestamp, and combines this with the current natural cooling capacity decay to determine the remaining available refrigerant storage capacity.

[0049] Calculate the shortest time span from the current moment to the start time of the peak-valley cooling storage interval. Based on the predicted total heat dissipation demand for each time period from the current moment to the start time of the next cycle, sum up the cumulative heat dissipation demand. Compare the remaining available cooling storage capacity with the cumulative heat dissipation demand to calculate the cooling storage gap.

[0050] If the cold storage shortage is greater than 0, it is determined that a cold storage shortage exists, and the temporary cold storage area construction process is initiated:

[0051] Based on the cold storage gap, combined with the cold storage efficiency of the refrigerant and the rated cooling power of the refrigeration unit, the temporary cold storage duration required to meet the cold storage gap is calculated.

[0052] Retrieve electricity price forecast data and forecast fluctuation range from the current time to the predicted start time of the next cycle's cold storage interval, set a temporary off-peak price threshold, and filter out the periods when the lower limit of the predicted electricity price fluctuation range is lower than the temporary off-peak price threshold and the upper limit is not higher than the temporary off-peak price threshold. Combined with the duration of temporary cold storage, define the temporary cold storage interval.

[0053] Specifically, the steps for generating the next cycle's cold storage capacity correction value include:

[0054] When the current cycle's peak-valley cold storage interval begins, the initial cold storage capacity carried over from the previous cycle to the current cycle and the actual cold storage consumption of the energy storage battery packs in the current cycle are retrieved. At the same time, if there is a temporary cold storage interval construction operation in the current cycle, the temporary cold storage capacity generated by the temporary cold storage interval is retrieved simultaneously to calculate the actual remaining cold storage capacity at the end of the current cycle.

[0055] Retrieve the baseline cold storage capacity for the current period and the predicted heat dissipation demand of the energy storage battery pack; and combine this with the total natural cooling capacity decay for the current period to obtain the theoretical remaining cold storage capacity for the current period.

[0056] By combining the actual remaining cold storage capacity with the theoretical remaining cold storage capacity, the deviation value of the cold storage capacity and the percentage of the cold storage capacity deviation are calculated; and the percentage of the cold storage capacity deviation is converted into a basic correction coefficient.

[0057] Retrieve the duration of the electricity price peak-valley cold storage interval for the next historical matching cycle and calculate the duration fluctuation range to calibrate the basic correction coefficient and generate the cold storage capacity correction coefficient.

[0058] The appropriate cooling capacity is calculated by retrieving the predicted total heat dissipation demand of the energy storage battery pack for the next cycle and the predicted threshold for heat dissipation demand. Combined with the cooling capacity correction coefficient, the corrected value for the cooling capacity for the next cycle is generated.

[0059] A method for controlling cold storage capacity based on real-time electricity prices includes:

[0060] The system acquires and periodically stores real-time electricity price data from the power grid, extracts historical periodic electricity price change characteristics, filters historical matching periods, and combines the current period's collected electricity price data change trend with the weighted electricity price data of historical matching periods to generate the current period's predicted electricity price data and the predicted electricity price fluctuation range.

[0061] The system collects operational data of the energy storage battery pack and, based on historical operational data from the historical fitting period, combines the charging and discharging plan parameters of the pack for the next period to predict the total heat dissipation demand of the energy storage battery pack for the next period and the predicted threshold for heat dissipation demand.

[0062] Based on the refrigerant storage efficiency and rated cooling power, the appropriate storage capacity and corresponding storage duration are calculated. Combined with the current cycle's predicted electricity price data and the predicted fluctuation range of electricity price, the effective candidate storage range is divided. Based on the benchmark storage duration, the effective candidate storage range is corrected, and the current cycle's peak-valley storage range is divided.

[0063] Predict the range of the peak-valley cold storage interval for the next electricity price cycle, and combine it with real-time monitoring of the refrigerant cold storage status to determine whether there is a cold storage gap. If so, construct a temporary cold storage interval.

[0064] Based on the current cycle's cold storage margin deviation, a correction value for the next cycle's cold storage capacity is generated.

[0065] The beneficial effects of this invention are:

[0066] This application uses a price cycle prediction module to periodically store real-time electricity price data from the power grid, extract historical cycle price change characteristics, and filter historical matching cycles. It then combines the current cycle price trend with weighted data from historical matching cycles to generate electricity price prediction data and fluctuation ranges, effectively improving the accuracy and reliability of electricity price prediction and avoiding increased operating costs due to misjudging the timing of cold storage caused by short-term price fluctuations. The demand prediction module collects operating data from energy storage battery packs and constructs a multi-condition heat dissipation demand fitting model based on historical fitting cycle data. Combined with the next cycle's charge / discharge plan parameters, it predicts heat dissipation demand and thresholds, accurately matching the actual heat dissipation demand under different operating conditions and solving the problem of cold energy supply and demand imbalance caused by traditional fixed-coefficient calculations. The cold storage zone division module calculates the baseline and suitable cold storage duration based on the refrigerant's cold storage efficiency and rated cooling power. It then selects and adjusts effective candidate cold storage zones based on the predicted fluctuation range of electricity prices to determine the peak-valley cold storage zone. This ensures that the cold storage capacity meets the heat dissipation demand of the next cycle without causing energy waste, significantly improving the profit margin of peak-valley electricity price arbitrage. At the same time, the monitoring and adjustment module predicts the range of the peak-valley cold storage zone for the next cycle, monitors the refrigerant cold storage status in real time, and constructs a temporary cold storage zone when a cold storage gap occurs. It can also generate a correction value for the cold storage capacity of the next cycle based on the current cycle's cold storage margin deviation, realizing dynamic closed-loop adjustment of the cold storage strategy. This effectively avoids situations of excess or insufficient cold storage and ensures the long-term stable and efficient operation of the energy storage system. Attached Figure Description

[0067] Figure 1 This is a flowchart of the cold storage control system based on real-time electricity pricing according to the present invention;

[0068] Figure 2 This is a flowchart illustrating the process of filtering historical matching periods in this invention;

[0069] Figure 3 This is a flowchart illustrating the process of dividing valid candidate cold storage zones according to the present invention;

[0070] Figure 4 A flowchart illustrating the construction of a temporary cold storage area for this invention. Detailed Implementation

[0071] Please see Figure 1 This embodiment introduces a cold storage control system based on real-time electricity prices, including an electricity price cycle prediction module, a demand prediction module, a cold storage zone division module, and a monitoring and adjustment module.

[0072] The electricity price cycle prediction module is used to acquire real-time electricity price data from the power grid and generate predicted electricity price data and the predicted fluctuation range for the current cycle based on historical cycle electricity price change characteristics. Specifically, it collects electricity price data within a preset constant cycle in real time through the power grid trading platform interface or wireless communication. The constant cycle is a fixed time length set by the system. The electricity price data collected in the current cycle is compared with historical electricity price data of the same type of cycle to extract and match the electricity price change characteristics. The historical electricity price data of the same type of cycle includes electricity price data of the same period and the same working day. The extracted features include electricity price peak value, valley value, peak-valley transition time, and electricity price change slope. Through time series prediction algorithm, combined with the matching degree between the current cycle electricity price change trend and the historical cycle, the electricity price value for the remaining time of the current cycle is predicted, and the predicted fluctuation range of the current electricity price is calculated based on the historical cycle electricity price prediction deviation rate.

[0073] In this embodiment, the core of the electricity price cycle prediction module is to solve the problems of insufficient accuracy, poor cycle matching, and lack of risk prediction in existing technologies. Existing technologies mostly rely on the real-time electricity price of a single cycle for simple peak-valley division, without combining the electricity price change patterns of the same type of cycle in history. This makes it impossible to capture the inherent characteristics of electricity price changes, and it is easy to misjudge peak and valley periods due to short-term electricity price fluctuations, which in turn leads to deviations in the selection of cold storage timing. At the same time, existing prediction schemes do not provide the predicted fluctuation range, and cannot quantify the prediction risk. This may cause the cold storage strategy based on the predicted electricity price to fail when the actual electricity price exceeds expectations, resulting in problems such as cold storage costs being higher than expected or insufficient cold storage capacity. This application ensures the accuracy and timeliness of real-time electricity price collection through the official power grid interface. By matching the characteristics of historical data of the same period, the electricity price forecast is made more consistent with the actual operating conditions and the electricity price change pattern, thus improving the reliability of the forecast trend. The time series forecasting algorithm can accurately predict the electricity price value for the remaining time of the current period. Combined with the historical forecast deviation rate, the fluctuation range is defined, which can provide a dual basis for the subsequent division of cold storage ranges, namely the predicted electricity price data and the risk boundary. This effectively avoids the failure of cold storage strategy due to electricity price forecast deviation, and ensures that the timing of cold storage is accurately matched with the electricity price valley range, laying the foundation for improving peak-valley arbitrage profits.

[0074] Preferably, the specific steps for generating the current period's predicted electricity price data and the predicted fluctuation range of electricity prices include:

[0075] Please see Figure 2The system acquires real-time electricity price data from the power grid and marks the collection timestamp. Based on a preset constant period, it periodically stores the electricity price data, constructs an electricity price database, and divides the data into a current period data storage partition and a historical period data storage partition. The current period data storage partition is used to update and store the electricity price data collected in the current constant period in real time, while the historical period data storage partition is used to retain the electricity price data of previous constant periods for a long time, providing data support for subsequent data filtering and feature extraction.

[0076] Historical electricity price data is retrieved from the historical periodic data storage partition of the electricity price database. The historical electricity price data is then double-filtered based on periodic time attributes and working day attributes, selecting historical electricity price data that are within the same period and working day type as the historical benchmark data. Electricity price change features are extracted from the selected historical benchmark data. The extraction dimensions include peak price features, valley price features, peak-valley transition time features, and price change slope features. The price change slope feature is calculated by the price difference and time interval between adjacent data collection times. The extracted electricity price change features are then correlated and integrated according to the periodic dimension, establishing an independent set of electricity price change features for each historical period of the same type. The feature sets of all historical periods of the same type of electricity price change feature are then aggregated to form a historical periodic electricity price change feature library.

[0077] The system retrieves collected electricity price data from the current period's data storage partition of the electricity price database. Following the same dimensions and methods as historical benchmark data, it extracts the electricity price change features for the current period's collected time period. Using a feature similarity algorithm, it compares each dimension of the current period's collected electricity price change features with the corresponding dimension features for each historical period in the historical electricity price change feature database for the same period, calculating the similarity coefficient between the current period's collected electricity price change features and the historical period's electricity price change features. Historical periods with similarity coefficients greater than a preset similarity threshold are selected as historical matching periods, and the average similarity coefficient of these historical matching periods is calculated as an indicator of the fit between the current period's electricity price change trend and the historical similar period's electricity price change trend. Specifically, the similarity threshold is a critical value used to determine whether the historical period's electricity price change features and the current period's electricity price change features have matching value. This critical value is determined by statistically analyzing the fit between the historical similar period's electricity price change features and the corresponding period's actual electricity price change trend, combined with the system's preset electricity price prediction accuracy target.

[0078] The electricity price data collected in the current period, the corresponding electricity price data from each historical matching period, and the fit index are correlated and integrated to form a prediction input dataset. The electricity price data from each historical matching period in the prediction input dataset are then weighted, with the weighting coefficient being the ratio of the similarity coefficient to the fit index for each historical matching period, resulting in weighted historical electricity price data. Based on the changing trend of the electricity price data collected in the current period, combined with the weighted historical electricity price data, a time-series data extrapolation method is used to extrapolate the electricity price values ​​for the remaining time in the current period, generating an initial electricity price prediction sequence for the remaining time in the current period. The time-series data extrapolation method refers to a calculation method that, based on the changing slope and characteristic trend of the electricity price data collected in the current period, combined with the changing patterns of the weighted historical electricity price data in the corresponding period, extrapolates the electricity price values ​​for the remaining time in the current period by following the current period's electricity price change trend and overlaying the historical matching period's electricity price change characteristics.

[0079] Calculate the deviation correction coefficient between the actual electricity price data for the corresponding prediction period in the historical matching cycle and the predicted electricity price data in the initial electricity price prediction sequence corresponding to the historical matching cycle. The deviation correction coefficient is the ratio of the actual electricity price to the predicted electricity price data in the historical matching cycle. Calculate the weighted average of the deviation correction coefficients for each historical matching cycle according to the similarity coefficient to obtain the comprehensive deviation correction coefficient. Multiply each predicted value in the initial electricity price prediction sequence by the comprehensive deviation correction coefficient to correct the deviation in the initial electricity price prediction sequence, forming the electricity price prediction sequence for the remaining time in the current cycle.

[0080] The predicted electricity price data and corresponding actual electricity price data for all historical matching periods are retrieved from the electricity price database. The absolute deviation between the predicted electricity price data and the corresponding actual electricity price data for each historical matching period is calculated for each time period. The relative deviation rate for each time period of each historical matching period is obtained by the ratio of the absolute deviation to the actual electricity price for the corresponding time period. The maximum value and the average value of the relative deviation rate for each time period of all historical matching periods are calculated. The maximum value is determined as the upper limit parameter of the deviation rate, and the average value is determined as the benchmark parameter of the deviation rate, which serves as the core basis for calculating the fluctuation range.

[0081] Each predicted value in the electricity price forecast sequence is used as a benchmark electricity value. A basic fluctuation threshold corresponding to each benchmark electricity value is calculated based on the deviation rate benchmark parameter. The basic fluctuation threshold is the product of the benchmark electricity value and the deviation rate benchmark parameter. The basic fluctuation threshold is verified in conjunction with the deviation rate upper limit parameter. First, the initial upper and lower boundaries of fluctuation corresponding to the benchmark electricity value are calculated based on the basic fluctuation threshold. Simultaneously, the maximum allowable fluctuation threshold is calculated using the product of the benchmark electricity value and the deviation rate upper limit parameter, and the upper and lower boundaries of the extreme fluctuation are determined. If the initial upper and lower boundaries of fluctuation exceed the range defined by the extreme fluctuation boundaries, the target fluctuation threshold is recalculated using the product of the deviation rate upper limit parameter and the benchmark electricity value. If it does not exceed the range, the basic fluctuation threshold is directly determined as the target fluctuation threshold. Using each benchmark electricity value as the center and the finally determined fluctuation threshold as the upper and lower boundaries, the electricity price forecast fluctuation range for each forecast time is defined. The fluctuation ranges for all forecast times are integrated to form a complete set of electricity price forecast fluctuation ranges for the remaining time of the current cycle.

[0082] The demand forecasting module is used to predict the total heat dissipation demand of the energy storage battery packs in the next cycle, providing an accurate cooling demand benchmark for the cold storage zone division module. Specifically, it collects real-time operating data of the energy storage battery packs, including pack charging and discharging power, operating time, ambient temperature, and corresponding real-time heat dissipation demand. The collected operating data is classified and archived according to a preset constant cycle to build a pack heat dissipation demand database covering multiple cycles. Based on historical data in the heat dissipation demand database, a correlation curve is fitted to show the changes in pack heat dissipation demand with charging and discharging power and ambient temperature under different operating conditions, clarifying the periodic variation law of the pack heat dissipation coefficient. Combining the operating data of the current cycle's current cycle's running time, the pack charging and discharging plan for the next cycle, and the ambient temperature prediction data, the fitted correlation curve is substituted to calculate the total heat dissipation demand of the packs in the next cycle. At the same time, the heat dissipation demand prediction threshold is determined based on the fluctuation range of historical cycle heat dissipation demand in the heat dissipation demand database.

[0083] In this embodiment, the core of the demand forecasting module is to address the problems of insufficient accuracy in forecasting heat dissipation demand for storage tanks and poor adaptability to electricity price cycles in existing technologies. Existing technologies mostly use a fixed heat dissipation coefficient to calculate heat dissipation demand, without considering fluctuations in charging and discharging power, changes in ambient temperature, and cyclical differences in storage tank operating conditions. This results in a large deviation between the predicted electricity price data and the actual value, leading to a mismatch between the cold storage capacity and the actual demand. This results in either excess cold storage leading to wasted energy or insufficient cold storage requiring peak-hour supplementary cooling. At the same time, existing forecasting schemes do not focus on the complete heat dissipation demand for the next cycle, and cannot provide accurate cold storage demand basis for the division of cold storage zones, affecting the implementation effect of peak-valley electricity price arbitrage strategies. This application constructs a multi-cycle heat dissipation demand database to accurately mine the heat dissipation patterns of the storage tanks. It dynamically adjusts the heat dissipation coefficient by combining charge and discharge plans with ambient temperature prediction data to improve the accuracy of heat dissipation demand prediction. By accurately predicting the total heat dissipation demand of the storage tanks in the next cycle, it ensures that the heat dissipation demand prediction results match the time sequence of the output data of the electricity price cycle prediction module. This provides two-way support for the division of cold storage zones based on both cooling demand and electricity price cycles, ensuring that the cold storage strategy meets both heat dissipation safety requirements and aligns with peak-valley electricity price arbitrage objectives.

[0084] Preferably, the specific steps for predicting the total heat dissipation demand of the energy storage battery pack in the next cycle include:

[0085] The system collects real-time operational data from the energy storage battery packs and marks each data point with a timestamp. The operational data is then partitioned and stored according to a preset constant cycle, creating a current cycle operational data storage partition and a historical cycle operational data storage partition. The current cycle operational data storage partition is used to update and store operational data from the battery packs during the current constant cycle, while the historical cycle operational data storage partition is used to retain complete operational data from previous constant cycles. Based on these partitioning rules, a structured battery pack heat dissipation requirement database is constructed to provide data support for subsequent data fitting and prediction calculations.

[0086] The number of fitting cycles is configured based on the stability characteristics of the battery pack's operating conditions. This configuration is used to retrieve historical operating data from the historical operating data storage partition of the heat dissipation demand database, representing the number of historical fitting cycles since the current cycle. Stability characteristics refer to the fluctuation, consistency, and historical reproducibility of core operating parameters affecting the heat dissipation demand of the battery pack within continuous historical operating cycles, such as charging / discharging mode, charging / discharging power fluctuation range, ambient temperature variation, battery pack operating cycle stage, and heat dissipation system efficiency. This is the core criterion for quantifying the reference fit of historical operating data to the current and next cycle's heat dissipation demand prediction, and thus determining the number of fitting cycles. For example, when the battery pack executes a fixed peak-valley arbitrage charging / discharging plan, if the charging / discharging power range, daily charging / discharging time period, and operating cycle stage are completely consistent across multiple consecutive cycles, the charging / discharging power fluctuation within a single cycle does not exceed ±5%, the ambient temperature variation range for the same time period does not exceed 3℃, and the heat dissipation system efficiency does not significantly decrease, then the operating condition exhibits high stability. The system allows for the configuration of more fitting periods, incorporating more historical data of the same type to improve the accuracy of heat dissipation demand fitting. When the energy storage battery pack executes grid dynamic frequency regulation and responsive charging and discharging plans, with frequent switching of charging and discharging power ranges and fluctuations exceeding ±20% within a single cycle, significant fluctuations in ambient temperature, or when the pack is in an unstable operating cycle stage of new commissioning or accelerated aging (i.e., the operating conditions are low-stability), fewer fitting periods are required to avoid fitting deviations caused by insufficient adaptability between long-term historical data and current operating conditions. The retrieved historical operating data is classified according to charging and discharging power range and ambient temperature range. For the historical operating data within each category range, the correlation coefficient between the pack's heat dissipation demand and charging and discharging power and ambient temperature is calculated using a data fitting algorithm, generating correlation curves of the pack's heat dissipation demand changing with charging and discharging power and ambient temperature under different operating conditions. The heat dissipation coefficient under the same operating conditions in each historical fitting period is statistically analyzed to clarify the pattern of change of the pack's heat dissipation coefficient with the operating cycle, forming a heat dissipation demand fitting model covering all operating conditions. Same operating conditions refer to operating conditions where the charging and discharging power range, ambient temperature range, and operating cycle stages of the charging box are all consistent.

[0087] The system retrieves the operating data of the current cycle's charging and discharging phases to obtain the charging and discharging plan parameters for the next cycle's charging and discharging phases. These parameters include the set values ​​of charging and discharging power for each time period and ambient temperature data. The system also obtains ambient temperature data for each time period from the meteorological forecasting platform and integrates the operating data of the current cycle's charging and discharging phases with the charging and discharging plan parameters for the next cycle to form a set of predictive input parameters.

[0088] Substitute the charging and discharging power settings and ambient temperature data for each period of the next cycle from the predicted input parameter set into the fitted correlation curve to calculate the heat dissipation demand of the charging box for each period of the next cycle; sum them up to obtain the predicted value of the total heat dissipation demand of the charging box in the next cycle.

[0089] Retrieve the predicted and actual values ​​of heat dissipation demand under the same operating conditions in the historical fitting period from the heat dissipation demand database, and calculate the deviation rate between the predicted and actual values ​​of heat dissipation demand for each period. Statistically calculate the deviation rate for all historical fitting periods, and determine the average and maximum deviation rates. Use the average value as the baseline deviation rate for heat dissipation demand and the maximum value as the upper limit of the deviation rate for heat dissipation demand. Based on the total heat dissipation demand of the next period, calculate the product of the total heat dissipation demand, the baseline deviation rate, and the upper limit of the deviation rate to obtain the upper and lower boundaries of the heat dissipation demand prediction threshold, thus forming a complete heat dissipation demand prediction result.

[0090] The cold storage zone division module is used to divide the current cycle's peak-valley cold storage zone based on the current cycle's predicted electricity price data and predicted price fluctuation range output by the electricity price cycle prediction module, and the next cycle's total heat dissipation demand and heat dissipation demand prediction threshold output by the demand prediction module. This provides a precise time period basis for cold storage control. Specifically, it first calculates the baseline cold storage capacity and corresponding baseline cold storage duration required to meet the heat dissipation demand based on the next cycle's total heat dissipation demand and refrigerant cold storage efficiency. Combining the current cycle's predicted electricity price data and predicted price fluctuation range, it selects periods where the lower limit of the predicted price fluctuation range is lower than the preset valley price threshold as basic candidate periods, and then eliminates periods where the upper limit of the predicted price fluctuation range is higher than the preset valley price threshold. The effective candidate time periods are obtained by setting a preset valley price threshold. These effective candidate time periods are then continuously integrated to form several consecutive effective candidate cold storage intervals. Based on the benchmark cold storage duration, the duration of the consecutive effective candidate cold storage intervals is verified and adjusted to ensure that the cold storage capacity that can be stored in the adjusted interval is not less than the total heat dissipation demand of the next cycle, and does not exceed the upper limit of the heat dissipation demand prediction threshold. This determines the peak-valley cold storage interval for the current cycle. If there is no consecutive effective candidate cold storage interval that meets the benchmark cold storage duration in the current cycle, non-consecutive effective candidate time periods with a low lower limit of the predicted electricity price fluctuation range are selected and their durations are accumulated until the accumulated duration reaches the benchmark cold storage duration, forming a supplementary cold storage interval.

[0091] In this embodiment, the core of the cold storage zone division module is to address the problems in existing technologies where cold storage zone division is not linked to the risk of electricity price fluctuations and has a low degree of matching with cooling demand. Existing technologies mostly divide cold storage zones based solely on electricity price forecasts, without considering the fluctuation range of electricity price forecasts. This can easily lead to situations where the actual electricity price exceeds the preset off-peak price threshold, resulting in cold storage costs being higher than expected. At the same time, existing technologies do not combine the actual heat dissipation demand of the next cycle to calculate the required cold storage duration, resulting in the divided cold storage zones either having insufficient duration to meet heat dissipation requirements or exceeding the requirements, leading to energy waste during off-peak hours. This application uses the electricity price fluctuation range as the core basis for selecting candidate time periods, ensuring that the selected time periods have an electricity price advantage while avoiding the cost risks caused by electricity price fluctuations. By linking with the cooling demand data of the demand forecast module, it achieves a precise match between the cold storage duration and the heat dissipation demand of the next cycle, ensuring that the cold storage strategy meets both heat dissipation safety requirements and aligns with the peak-valley electricity price arbitrage objective.

[0092] Preferably, the specific steps for dividing the current cycle's electricity price peak-valley cold storage range include:

[0093] Please see Figure 3 The system acquires real-time electricity price data from the power grid and predicted electricity price data for the current period, generating an electricity price sequence for the current period. The stability of the current period's electricity price is measured by the average deviation rate between the predicted and actual electricity prices for the collected time periods. If the average deviation rate is lower than a preset stability threshold, it indicates that the current period's electricity price prediction is trending towards stability, providing a data basis for cold storage zone division, and the current period's electricity price sequence is determined to be stable. Otherwise, the current period's electricity price sequence is determined to be unstable, and a new determination is made after the electricity price prediction data is updated. The stability threshold is a critical value used to determine whether the accuracy of the electricity price prediction data meets the requirements for cold storage zone division, and it is set by statistically analyzing the probability distribution of the average deviation rate of historical period electricity price predictions.

[0094] The cold storage zone trigger time is set based on the remaining time of the current cycle as the time threshold reserved for the pre-start of the refrigerant chiller unit. This time is the critical moment for starting the peak-valley cold storage zone division operation at the latest. If the cold storage zone trigger time is reached or the current cycle's electricity price sequence is determined to be stable, the peak-valley cold storage zone division operation will be started.

[0095] Based on the total heat dissipation demand of the next cycle's refrigerant-based chiller and the predicted heat dissipation demand threshold, combined with the system's preset refrigerant storage efficiency, the required target cooling capacity is calculated by the ratio of the total heat dissipation demand of the next cycle's refrigerant-based chiller to the refrigerant storage efficiency. The refrigerant storage efficiency refers to the ratio of the effective cooling capacity actually stored by the refrigerant chiller unit to the theoretical cooling capacity corresponding to the unit's consumed electrical energy. Based on historical operating data of the refrigerant chiller unit under different charge / discharge power and ambient temperature conditions, the average efficiency under each condition is calculated, combined with the current cycle's ambient temperature prediction data and the chiller unit's operating plan settings.

[0096] Based on the rated cooling power of the refrigerant refrigeration unit and the target cold storage capacity, the baseline cold storage duration required to meet the target cold storage capacity is calculated. Based on the upper limit of the heat dissipation demand prediction threshold, the appropriate cold storage capacity is calculated. Combined with the rated cooling power of the refrigerant refrigeration unit, the appropriate cold storage duration is obtained, forming a reasonable range for the cold storage duration.

[0097] The system receives the current cycle's predicted electricity price data and the predicted price fluctuation range, and sets a preset valley price threshold. It then iterates through each remaining time period of the current cycle, selecting periods where the lower limit of the predicted price fluctuation range is lower than the preset valley price threshold as basic candidate periods. The preset valley price threshold refers to the critical electricity price value that defines the valley price period. It is set by statistically analyzing the historical cycle's valley price range and combining it with the system's peak-valley electricity price arbitrage profit target.

[0098] All basic candidate time periods are sorted in chronological order, and adjacent basic candidate time periods are integrated to form several consecutive effective candidate cold storage intervals; the duration of each consecutive effective candidate cold storage interval is calculated and the duration information is labeled.

[0099] The duration of each consecutive valid candidate cold storage interval is compared with the reasonable range of cold storage duration. If there is a consecutive valid candidate cold storage interval whose duration is within the reasonable range of cold storage duration, the valid candidate cold storage interval with the smallest deviation from the benchmark cold storage duration is selected as the electricity price peak-valley cold storage interval for the current cycle.

[0100] If the duration of consecutive valid candidate cold storage intervals exceeds the appropriate cold storage duration, the interval will be truncated, and the period starting from the lowest point of the lower limit of the electricity price prediction fluctuation range and having a duration equal to the appropriate cold storage duration will be retained as the electricity price peak-valley cold storage interval.

[0101] If the duration of all consecutive valid candidate cold storage intervals in the current period does not reach the benchmark cold storage duration, then all valid candidate periods are sorted from low to high according to the lower limit of the electricity price prediction fluctuation range; the duration of the first-ranked valid candidate periods is accumulated until the accumulated duration reaches the benchmark cold storage duration, and the selected multiple non-consecutive valid candidate periods are integrated into the electricity price peak-valley cold storage interval.

[0102] The start and end times and durations of the determined peak-valley cold storage intervals or supplementary cold storage intervals are compiled to form the cold storage interval plan for the current period.

[0103] The monitoring and adjustment module is used to comprehensively measure the peak-valley cooling storage range of the electricity price in the next historical matching cycle, generate the predicted range of the peak-valley cooling storage range of the next cycle, and monitor the cooling storage status of the refrigerant in real time to determine whether there is a cooling storage gap. If so, a temporary cooling storage range is constructed. At the same time, based on the current cycle's cooling storage margin deviation, a cooling storage correction value for the next cycle is generated to realize dynamic closed-loop adjustment of the cooling storage capacity, ensuring continuous coverage of heat dissipation demand and efficient utilization of cooling storage resources. Specifically, the system comprehensively measures the peak-valley electricity price range for the next historical matching cycle, using multi-dimensional indicators to extract the variation patterns and generate a predicted range for the next cycle. It also collects real-time status data such as the actual cooling capacity and cooling capacity decay rate of the refrigerant cooling system, combining this data with the predicted range and the heat dissipation demand data from the demand forecasting module to assess the support capacity of the remaining cooling capacity. For scenarios with insufficient cooling capacity, it selects temporary cooling periods and calculates the required cooling capacity to construct temporary cooling ranges. Finally, based on factors such as the remaining cooling capacity deviation, temporary cooling consumption, and equipment efficiency fluctuations, it generates a cooling capacity correction value for the next cycle, which is used to adjust the cooling capacity adapted to the next cycle in the cooling range division module, achieving dynamic optimization of the cooling strategy.

[0104] In this embodiment, the core of the monitoring and adjustment module is to address the problems in existing technologies, such as the lack of predictive capability for cold storage zones, the absence of dynamic assessment of cold storage capacity, and the lack of a closed-loop correction mechanism for cold storage capacity. Existing technologies only execute static cold storage strategies for the current cycle's cold storage zone, failing to predict the range of the next cycle based on historical matching cycle cold storage zone characteristics. Furthermore, they do not monitor changes in cold storage capacity in real time, making it prone to issues like cold storage zone shifts and cooling capacity decay leading to insufficient cold storage capacity to support the next cycle's cold storage zone, or excessive cold storage resulting in resource waste. Simultaneously, existing solutions lack a cold storage capacity correction mechanism, making it impossible to adjust subsequent strategies based on previous cold storage deviations, leading to recurring supply-demand imbalances. This application achieves accurate prediction of the next cycle's range through comprehensive measurement of historical matching cycle cold storage zones, providing a forward-looking basis for cold storage capacity assessment. It mitigates the risk of heat dissipation gaps through real-time monitoring and temporary cold storage replenishment. A closed-loop correction mechanism generates a corrected value for the next cycle's cold storage capacity, dynamically adapting to changes in the cold storage zone and fluctuations in cooling demand, ensuring that the cold storage strategy is both safe and economical.

[0105] Please see Figure 4 Preferably, the specific steps for constructing a temporary cold storage area include:

[0106] The electricity price cycle prediction module filters historical matching cycles for the next cycle based on the cycle time period attributes, working day attributes, and electricity price change characteristics. It then retrieves data on peak-valley cooling storage interval schemes for historical matching cycles, including start and end times, interval duration, and suitable cooling storage capacity. Specifically, based on the cycle time period attributes and working day attributes of the next cycle, it focuses on aligning the electricity price change characteristics of the next cycle with historical cycles. A feature similarity algorithm is used to compare the peak-valley distribution, peak-valley transition time, and electricity price change slope characteristics of each historical cycle with the next cycle, eliminating historical cycles with significant deviations in electricity price change trends. This ensures that the selected historical matching cycles accurately reflect the electricity price fluctuation patterns of the next cycle, providing a reliable reference for subsequent interval predictions.

[0107] Clustering algorithms are called to perform clustering operations on the start and end times of the electricity price peak-valley cold storage interval for all historical matching cycles, generating corresponding start time clusters and end time clusters; the data centrality within the cluster is calculated, and the start time cluster and end time cluster with the highest data centrality within the cluster are selected as the core start time cluster and core end time cluster.

[0108] Based on the time distribution of the core start time cluster and the core end time cluster, extract all time data within the core start time cluster, calculate the maximum and minimum start time values, and generate the start time range of the electricity price peak-valley cold storage interval for the next cycle; process the core end time cluster in the same way to generate the end time range of the electricity price peak-valley cold storage interval for the next cycle.

[0109] Continuously collect real-time refrigerant storage capacity and mark the collection timestamp; combine this with the current natural cooling capacity decay to determine the remaining available refrigerant storage capacity; calculate the shortest time span from the current moment to the start time of the peak-valley electricity price storage interval; based on the demand forecasting module, predict the total heat dissipation demand for each time period from the current moment to the predicted start time of the next cycle's storage interval, accumulate these values ​​to obtain the cumulative heat dissipation demand, compare the remaining available refrigerant storage capacity with the cumulative heat dissipation demand, calculate the storage gap; if the storage gap is greater than 0, it is determined that a storage gap exists, and the temporary storage interval construction process is initiated: where natural cooling capacity decay refers to the refrigerant storage system... The system measures the amount of cooling loss caused by factors such as ambient temperature conduction and natural heat dissipation of the cooling medium in the absence of active cooling and heat release operations. This is obtained by collecting the natural cooling decay rate per unit time in real time, multiplying it by the duration from the current moment to the cooling decay calculation moment, and then verifying it by combining the difference between the initial cooling capacity and the real-time cooling capacity. Specifically, for the total heat dissipation demand in each time period from the current moment to the predicted start time of the next cycle's cooling storage interval, a data retrieval request is sent to the demand prediction module to extract the predicted heat dissipation demand value of the energy storage battery pack in that time period. After accumulating the predicted values ​​of each time period, the cumulative heat dissipation demand is obtained.

[0110] Based on the cold storage gap, combined with the refrigerant's cold storage efficiency and the rated cooling capacity of the refrigeration unit, the temporary cold storage duration required to meet the gap is calculated. Electricity price forecast data and predicted fluctuation ranges from the current time to the predicted start time of the next cycle's cold storage interval are retrieved. A temporary valley price threshold is set based on the historical cycle's electricity price valley range and the system's peak-valley arbitrage profit target. Periods where the lower limit of the predicted electricity price fluctuation range is lower than the temporary valley price threshold and the upper limit is not higher than the temporary valley price threshold are selected. These selected periods are integrated in chronological order, and the combination of periods with a continuous duration equal to or closest to the temporary cold storage duration is defined as the temporary cold storage interval, with the start and end times of the temporary cold storage interval clearly defined. Specifically, the electricity price forecast data and predicted fluctuation ranges from the current time to the predicted start time of the next cycle's cold storage interval are obtained by sending a targeted data retrieval command to the electricity price cycle forecast module. This integrates the electricity price forecast data and predicted fluctuation ranges for the remaining periods of the current cycle and the early periods of the next cycle within this time period, forming an electricity price forecast dataset covering the entire time period.

[0111] Organize the core parameters of the temporary cold storage zone, including start time, end time, zone duration, and required temporary cold storage capacity. Combine these with the refrigeration unit operating parameter specifications to form a standardized temporary cold storage plan and initiate the temporary cold storage operation. Continue until the cold storage capacity reaches the preset temporary cold storage capacity to complete the construction and execution of the temporary cold storage zone.

[0112] Preferably, the specific steps for generating the corrected value for the next cycle's cold storage capacity based on the current cycle's cold storage capacity deviation include:

[0113] When the current cycle's peak-valley cold storage interval begins, the initial cold storage capacity carried over from the previous cycle is retrieved, along with the actual cold storage consumption of the energy storage battery packs within the current cycle. If a temporary cold storage interval is constructed in the current cycle, the temporary cold storage capacity generated by the temporary cold storage interval is retrieved simultaneously. The initial cold storage capacity carried over from the previous cycle is added to the temporary cold storage capacity, and then the actual cold storage consumption of the current cycle is subtracted to obtain the actual remaining cold storage capacity at the end of the current cycle. The data collection timestamp and parameter source of each data point are simultaneously marked to ensure data traceability.

[0114] The system retrieves the baseline cold storage capacity for the current period from the cold storage zone division module and the predicted heat dissipation demand of the energy storage battery pack for the current period from the demand forecast module. Simultaneously, it collects the total natural cooling capacity decay generated by the refrigerant cold storage system within the current period. The system then adds the temporary cold storage capacity to the baseline cold storage capacity for the current period, and subtracts the predicted heat dissipation demand and the total natural cooling capacity decay to obtain the theoretical remaining cold storage capacity for the current period, thus completing the preparation of basic data for comparing the theoretical and actual values.

[0115] The actual remaining cold storage capacity for the current period is subtracted from the theoretical remaining cold storage capacity to obtain the cold storage capacity deviation value. The deviation attribute is determined based on the sign of the deviation value: a deviation value greater than zero indicates cold storage surplus, and the larger the deviation value, the higher the cold storage surplus; a deviation value less than zero indicates cold storage shortage, and the larger the absolute value of the deviation value, the higher the cold storage shortage. Simultaneously, the proportion of the cold storage capacity deviation value to the current period's suitable cold storage capacity is calculated to obtain the cold storage capacity deviation ratio. A basic correction coefficient is generated based on the cold storage capacity deviation ratio. The specific conversion logic is as follows: when cold storage is insufficient, the basic correction coefficient is greater than 1, and the higher the cold storage capacity deviation ratio, the higher the value of the basic correction coefficient; when cold storage is excessive, the basic correction coefficient is less than 1, and the higher the cold storage capacity deviation ratio, the lower the value of the basic correction coefficient.

[0116] The actual duration of the peak-valley cooling storage interval for all historical matching cycles in the next cycle is retrieved. Combined with the similarity coefficient of each historical matching cycle, the duration fluctuation range of the peak-valley cooling storage interval in the next cycle is calculated: First, the similarity coefficient of each historical matching cycle is used as a weighting factor to calculate the weighted average of the cooling storage interval duration for all historical matching cycles, obtaining the weighted benchmark duration for the cooling storage interval in the next cycle. Then, the relative deviation rate between the cooling storage interval duration of each historical matching cycle and the weighted benchmark duration is calculated cycle by cycle. The maximum value of all relative deviation rates is taken as the duration fluctuation range. Based on the probability distribution of the relative deviation rate of the duration in historical matching cycles, a three-level threshold for duration fluctuation range is defined. For example, the low fluctuation threshold is a fluctuation range less than or equal to 5%, the medium fluctuation threshold is a fluctuation range greater than 5% and less than or equal to 15%, and the high fluctuation threshold is a fluctuation range greater than 15%. Corresponding levels are matched for different fluctuation levels. The adjustment coefficient is linked to the deviation direction of the base correction coefficient: When the base correction coefficient is greater than 1, indicating insufficient cooling storage in the previous cycle, the low, medium, and high fluctuation ranges correspond to adjustment coefficients of 1.0, 1.08, and 1.15, respectively. The larger the fluctuation range, the higher the adjustment coefficient, amplifying the upward adjustment of cooling storage capacity to cope with the risk of range fluctuations. When the base correction coefficient is less than 1, indicating excess cooling storage in the previous cycle, the low, medium, and high fluctuation ranges correspond to adjustment coefficients of 1.0, 0.95, and 0.88, respectively. The larger the fluctuation range, the lower the adjustment coefficient, narrowing the downward adjustment of cooling storage capacity to avoid the risk of heat dissipation gaps. When the base correction coefficient is equal to 1, indicating no deviation in cooling storage capacity in the previous cycle, all fluctuation ranges correspond to an adjustment coefficient of 1.0. The base correction coefficient is multiplied by the matched adjustment coefficient to complete the calibration, ensuring that the corrected cooling storage capacity can adapt to the fluctuation range of electricity price and heat dissipation demand in the next cycle, generating the final cooling storage capacity correction coefficient.

[0117] The system retrieves the adapted cooling capacity from the cooling storage zone division module. This adaptation capacity is calculated based on the predicted total heat dissipation demand of the energy storage battery packs in the next cycle, the refrigerant cooling efficiency, and the upper limit of the predicted heat dissipation demand threshold. The adapted cooling capacity is then multiplied by the final cooling capacity correction coefficient to calculate the corrected target cooling capacity for the next cycle. Simultaneously, the difference between the corrected target cooling capacity and the original adapted cooling capacity is used to determine the corrected cooling capacity value for the next cycle. The system also retrieves the predicted heat dissipation demand threshold for the next cycle from the demand forecasting module and calculates the minimum guaranteed cooling capacity and the maximum limited cooling capacity corresponding to the upper and lower boundaries of this threshold, respectively, to adjust the cooling capacity. The rationality of the target cold storage capacity for the next cycle is verified: if the revised target cold storage capacity for the next cycle is lower than the minimum guaranteed cold storage capacity, the minimum guaranteed cold storage capacity will be used as the final target cold storage capacity for the next cycle; if the revised target cold storage capacity for the next cycle is higher than the maximum limited cold storage capacity, the maximum limited cold storage capacity will be used as the final target cold storage capacity for the next cycle; if the revised target cold storage capacity for the next cycle is between the minimum guaranteed cold storage capacity and the maximum limited cold storage capacity, the revised value will be used directly as the final target cold storage capacity for the next cycle, thus avoiding insufficient or excessive cold storage throughout the process.

[0118] This embodiment introduces a cold storage control method based on real-time electricity pricing, including:

[0119] Step S1: Collect real-time electricity price data of the power grid within a preset constant period through the power grid trading platform interface or wireless communication, mark the timestamps, and periodically store the data to build an electricity price database; retrieve historical electricity price data, filter historical benchmark data according to the period and working day attributes, extract electricity price change characteristics, and integrate them to build a historical electricity price change characteristic library of the same period; extract the electricity price characteristics of the current period's collected periods, compare and filter historical matching periods, and calculate the matching degree index; combine the current period's electricity price trend, historical matching period weighted data, and time series prediction algorithm to deduce the initial electricity price prediction sequence, and obtain the final sequence after deviation correction; calculate the relative deviation rate through historical prediction and actual electricity price data, determine the deviation rate parameter, calculate and verify the fluctuation threshold based on the electricity price prediction value, define the electricity price prediction fluctuation range at each prediction time, form a complete set of fluctuation ranges, and provide data and risk boundary support for the division of cold storage areas.

[0120] Step S2: Real-time collection of operational data such as charging and discharging power, operating time, ambient temperature, and real-time heat dissipation demand of the energy storage battery pack. After marking with timestamps, the data is periodically partitioned and stored to construct a multi-cycle structured pack heat dissipation demand database. Based on the stability of the pack's operating conditions, the number of fitting cycles is configured, and the corresponding historical fitting cycle data is retrieved. The data is classified according to the charging and discharging power and ambient temperature ranges. The heat dissipation demand correlation curves under different operating conditions are generated through data fitting algorithms, and the heat dissipation coefficient patterns are statistically analyzed to form a heat dissipation demand fitting model for all operating conditions. The current operating data, the next cycle charging and discharging plan, and the ambient temperature prediction data are integrated to calculate the heat dissipation demand for each time period in the next cycle and sum them to obtain the total predicted value. The baseline deviation rate and upper limit of the heat dissipation demand are determined through historical operating condition deviation analysis. Combined with the total demand, the upper and lower boundaries of the heat dissipation demand prediction threshold are calculated to form a complete heat dissipation demand prediction result.

[0121] Step S3: Obtain real-time electricity price data and current cycle predicted electricity price data to generate an electricity price sequence. Determine the stability of the electricity price by the average deviation rate between the predicted and actual values ​​of the collected time period. If the deviation rate is below the threshold, the price is considered stable; otherwise, wait for data updates. Simultaneously, set the trigger time for the cold storage interval based on the remaining duration of the current cycle. Initiate the division operation after the trigger time is reached or the price is determined to be stable. Calculate the target and suitable cold storage capacity based on the heat dissipation demand of the next cycle, the predicted threshold, and the cold storage efficiency of the refrigerant. Combine this with the rated cooling power of the chiller unit to obtain the benchmark cold storage duration and the suitable cold storage duration, forming a reasonable range for the cold storage duration. Set a valley price threshold and filter out basic candidate periods where the lower limit of the predicted electricity price fluctuation range is below the threshold. Integrate these into continuous and valid candidate intervals and label their durations. Determine the peak-valley cold storage interval for the current cycle through duration verification and adjustment, and organize the parameters to form a cold storage interval scheme.

[0122] Step S4: Based on the cycle period, working day attributes, and electricity price change characteristics of the next cycle, filter historical matching cycles and retrieve their electricity price peak-valley cooling storage interval scheme data; call the clustering algorithm to cluster the start and end times of the historical matching cycle intervals, generate time clusters and calculate the concentration, select the core time clusters to generate the start and end time ranges of the electricity price peak-valley cooling storage interval for the next cycle; continuously collect the real-time cooling capacity of the refrigerant and mark the timestamp, and calculate the remaining available cooling capacity by combining it with the natural cooling capacity decay; calculate the shortest time span from the current time to the start time of the predicted interval for the next cycle, accumulate the predicted heat dissipation demand for this period to obtain the cumulative demand, compare and calculate the cooling storage gap, if it is greater than 0, start the temporary cooling process; calculate the temporary cooling duration based on the gap, refrigerant cooling efficiency, and unit power, retrieve the corresponding electricity price prediction data, set the temporary valley price threshold to filter suitable time periods, integrate and delineate the temporary cooling storage interval, form a standardized scheme, and execute the cooling storage operation.

[0123] Step S5: Upon reaching the start time of the current cycle's peak-valley cooling storage interval, retrieve the initial cooling storage capacity carried over from the previous cycle and the actual cooling storage consumption for the current cycle. If temporary cooling storage exists, retrieve it simultaneously. Calculate the actual remaining cooling storage capacity for the current cycle and mark the data source. Retrieve the current cycle's baseline cooling storage capacity, predicted heat dissipation demand, and total natural cooling capacity decay to calculate the theoretical remaining cooling storage capacity. Compare the actual and theoretical values ​​to obtain the cooling storage capacity deviation value, determine the deviation attribute, calculate the cooling storage capacity deviation ratio, and generate a basic correction coefficient. Retrieve the interval duration of the historical matching cycle for the next cycle and calculate the fluctuation amplitude. Define the duration-level threshold matching adjustment coefficient and calibrate to obtain the final cooling storage capacity correction coefficient. Combine the cooling storage capacity adapted for the next cycle to determine the correction value, verify its rationality against the predicted heat dissipation demand threshold, provide a basis for parameter optimization, and achieve dynamic closed-loop adjustment of cooling storage capacity.

[0124] Working principle and its effects:

[0125] This invention utilizes the coordinated linkage of an electricity price cycle prediction module, a demand prediction module, a cold storage zone division module, and a monitoring and adjustment module. With real-time electricity prices and energy storage battery pack operation data as core inputs, it constructs a full-process cold storage control logic of prediction, division, monitoring, and correction. This enables the dynamic adaptation of cold storage strategies to electricity price fluctuations and heat dissipation demands, solving the problem of blind decision-making in traditional cold storage and improving the economy and stability of energy storage system operation.

[0126] This invention periodically stores real-time electricity price data from the power grid, extracts historical periodic electricity price change characteristics, and filters historical matching periods. It combines the current periodic electricity price trend with weighted historical data to generate predicted electricity prices and fluctuation ranges. At the same time, it optimizes prediction accuracy through deviation correction and threshold verification, effectively avoiding increased cooling costs or insufficient energy due to incorrect electricity price predictions. It collects and stores the operation data of energy storage battery packs in partitions, constructs multi-condition heat dissipation demand correlation curves based on historical fitted period data, and predicts heat dissipation demand and thresholds by combining the next period's charge and discharge plan with ambient temperature data. This accurately matches dynamically changing heat dissipation demand and solves the problem of cooling supply and demand imbalance caused by fixed coefficient calculations. Based on the calculation benchmarks of refrigerant storage efficiency and rated cooling power, and the appropriate storage duration, and combined with the predicted fluctuation range of electricity prices, effective candidate storage ranges are screened and corrected. Through duration verification, truncation, and cumulative adjustment, the optimal peak-valley storage range is determined to ensure that the storage capacity meets demand without causing energy waste, maximizing peak-valley arbitrage profits. The range of the peak-valley storage range for the next cycle is predicted through clustering algorithms, and the refrigerant storage status and natural cooling capacity decay are monitored in real time. When a storage gap occurs, a temporary storage range is quickly constructed. At the same time, based on the deviation of the current cycle's storage capacity, the correction coefficient is converted and calibrated to generate the correction value of the next cycle's storage capacity, realizing the dynamic closed-loop adjustment of the storage strategy.

[0127] In summary, this invention, through deep collaboration and refined control of various modules, organically integrates the accuracy of electricity price forecasting, the adaptability of heat dissipation requirements, the optimization of the cold storage area, and the dynamism of the adjustment mechanism to form a complete cold energy storage control system. This not only effectively reduces the operating cost of the energy storage system and ensures the safety of battery operation, but also continuously optimizes the cold storage strategy through a closed-loop correction mechanism, providing reliable support for the efficient operation of the energy storage system under the peak-valley electricity pricing mechanism.

[0128] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A cold storage control system based on real-time electricity pricing, characterized in that, It includes an electricity price cycle forecasting module, a demand forecasting module, a cold storage zone division module, and a monitoring and adjustment module; The electricity price cycle prediction module is used to acquire real-time electricity price data of the power grid and store it periodically, extract historical cycle electricity price change characteristics, filter historical matching cycles, and combine the current cycle's collected electricity price data change trend with the weighted electricity price data of historical matching cycles to generate the current cycle's predicted electricity price data and the predicted electricity price fluctuation range. The demand forecasting module is used to collect the operating data of the energy storage battery pack, and based on the historical operating data of the historical fitting cycle, combined with the charging and discharging plan parameters of the pack in the next cycle, to predict the predicted value of the total heat dissipation demand of the energy storage battery pack in the next cycle and the heat dissipation demand prediction threshold. The cold storage zone division module is used to calculate the appropriate cold storage capacity and corresponding appropriate cold storage duration based on the cold storage efficiency of the refrigerant and the rated cooling power. It combines the current cycle's predicted electricity price data and the predicted fluctuation range of electricity price to divide the effective candidate cold storage zones. Based on the benchmark cold storage duration, the effective candidate cold storage zones are corrected to divide the current cycle's peak-valley cold storage zones. The monitoring and adjustment module determines whether there is a cold storage gap by predicting the range of the peak-valley cold storage interval for the next electricity price cycle and combining it with real-time monitoring of the cold storage status of the refrigerant. If so, a temporary cold storage interval is constructed. At the same time, based on the current cycle's cold storage margin deviation, a correction value for the cold storage capacity in the next cycle is generated. The steps for dividing the effective candidate cold storage areas include: The system acquires real-time electricity price data and current cycle predicted electricity price data, generates the current cycle electricity price sequence, and calculates the average deviation rate between the predicted and actual electricity price values ​​for the current cycle's collected periods. If the average deviation rate is lower than a preset stability threshold, the current cycle electricity price sequence is determined to be stable. Set a trigger time for the cold storage zone. If the trigger time is reached or the current electricity price sequence is determined to be stable, then initiate the electricity price peak-valley cold storage zone division operation: Based on the total heat dissipation demand of the next cycle's insertion box and the preset refrigerant storage efficiency, the baseline storage capacity is calculated; based on the rated cooling power and the baseline storage capacity, the baseline storage duration is calculated. Based on the upper limit of the predicted heat dissipation demand threshold, the appropriate cold storage capacity is calculated, and combined with the rated cooling power, the appropriate cold storage duration is obtained, forming a reasonable range for the cold storage duration. Receive the current cycle's predicted electricity price data and the predicted electricity price fluctuation range, and set a preset valley price threshold; iterate through each time period of the remaining time in the current cycle, and filter out the time periods where the lower limit of the predicted electricity price fluctuation range is lower than the preset valley price threshold, as the basic candidate time periods; All basic candidate time periods are sorted in chronological order and integrated to form effective candidate cold storage intervals; the duration of each effective candidate cold storage interval is calculated and labeled with duration information. The steps for dividing the current cycle's electricity price peak-valley cold storage range include: The duration of each valid candidate cold storage interval is compared with the reasonable range of cold storage duration. If the duration of a valid candidate cold storage interval is within the reasonable range of cold storage duration, the valid candidate cold storage interval with the smallest deviation from the benchmark cold storage duration is selected as the peak-valley cold storage interval for the current cycle. If the duration of a valid candidate cold storage interval exceeds the appropriate cold storage duration, the valid candidate cold storage interval will be truncated, and the period starting from the lowest point of the lower limit of the electricity price prediction fluctuation range and having a duration equal to the appropriate cold storage duration will be retained as the electricity price peak-valley cold storage interval. If the duration of all valid candidate cold storage intervals in the current cycle does not reach the benchmark cold storage duration, then all valid candidate periods are sorted from low to high according to the lower limit of the electricity price prediction fluctuation range; the duration of each valid candidate period is accumulated until the accumulated duration reaches the benchmark cold storage duration, and the selected valid candidate periods are integrated into the electricity price peak-valley cold storage interval.

2. The cold storage control system based on real-time electricity pricing as described in claim 1, characterized in that, The steps for filtering historical matching periods include: Acquire real-time electricity price data from the power grid, mark the collection timestamp, periodically store the electricity price data according to a preset constant period, construct an electricity price database, and divide the data storage partition for the current period into a data storage partition for the historical period. Historical electricity price data is retrieved from the historical periodic data storage partition of the electricity price database. The historical electricity price data is then filtered based on the periodic time attribute and the working day attribute to obtain historical benchmark data. Electricity price change characteristics are extracted to construct a historical electricity price change feature library of the same period. Retrieve the collected electricity price data from the current period data storage partition of the electricity price database, and extract the electricity price change characteristics of the collected period in the current period; Calculate the similarity coefficient between the electricity price change characteristics of the current period and the electricity price change characteristics of each historical period; select historical periods with similarity coefficients greater than the preset similarity threshold as historical matching periods, and calculate the average of the similarity coefficients of historical matching periods as an indicator of the fit between the current period's electricity price change trend and the historical electricity price change trend of the same type of period.

3. The cold storage control system based on real-time electricity pricing as described in claim 2, characterized in that, The steps for generating the current period's predicted electricity price data and the predicted fluctuation range of electricity prices include: The electricity price data of the current period, the electricity price data of the corresponding period of each historical matching period, and the matching index are linked and integrated to form a predictive input dataset. The historical matching period electricity price data in the prediction input dataset are weighted, and the weighting coefficient is the ratio of the similarity coefficient to the fit index of each historical matching period, to obtain the weighted historical electricity price data. Based on the electricity price data collected for the current period, combined with weighted historical electricity price data, an initial electricity price prediction sequence for the remaining time periods of the current period is generated using a time-series data extrapolation method. Calculate the deviation correction coefficient between the actual electricity price data of the corresponding prediction period in the historical matching cycle and the predicted electricity price data of the initial electricity price prediction sequence corresponding to the historical matching cycle; calculate the deviation correction coefficient of each historical matching cycle by weighting the similarity coefficient to obtain the comprehensive deviation correction coefficient; correct the deviation of the initial electricity price prediction sequence to form the electricity price prediction sequence for the remaining time of the current cycle. Retrieve the predicted electricity price data and the corresponding actual electricity price data for all historical matching periods, and calculate the absolute deviation between the predicted electricity price data and the corresponding actual electricity price data for each historical matching period. By the ratio of the absolute deviation to the actual electricity price for the corresponding period, the relative deviation rate for each period of each historical matching period is obtained. The maximum and average relative deviation rates for each time period in all historical matching cycles are statistically analyzed. The maximum value is used as the upper limit parameter of the deviation rate, and the average value is used as the benchmark parameter of the deviation rate. Each predicted electricity price in the electricity price prediction sequence is used as a benchmark electricity value; the basic fluctuation threshold corresponding to each benchmark electricity value is calculated based on the deviation rate benchmark parameter, and the basic fluctuation threshold is verified by combining the deviation rate upper limit parameter to obtain the target fluctuation threshold, which is used to define the electricity price prediction fluctuation range at each prediction time.

4. The cold storage control system based on real-time electricity pricing as described in claim 1, characterized in that, The steps of predicting the total heat dissipation demand of the energy storage battery pack for the next cycle and the heat dissipation demand prediction threshold include: Real-time acquisition of operational data from energy storage battery packs, marking the acquisition timestamp; and partitioning the operational data into current cycle operational data storage partitions and historical cycle operational data storage partitions; Configure the number of fitting cycles, which is used to retrieve historical running data of the number of historical fitting cycles that are the number of fitting cycles away from the current cycle from the historical running data storage partition of the heat dissipation demand database; The retrieved historical operating data are classified according to the charging and discharging power range and the ambient temperature range. For the historical operating data in each category range, the correlation coefficient between the heat dissipation requirement of the charging box and the charging and discharging power and the ambient temperature is calculated, and the correlation curve of the heat dissipation requirement of the charging box with the charging and discharging power and the ambient temperature under different operating conditions is generated. Retrieve the running data of the current cycle's plug-in box during the running period, obtain the charging and discharging plan parameters of the plug-in box for the next cycle, and form a set of predictive input parameters. The charging and discharging plan parameters include the set values ​​of charging and discharging power for each time period and ambient temperature data. Substitute the charging and discharging power setpoints and ambient temperature data for each period of the next cycle from the predicted input parameter set into the fitted correlation curve to calculate the heat dissipation demand of the charging box for each period of the next cycle; sum them up to obtain the predicted value of the total heat dissipation demand of the charging box in the next cycle. Retrieve the predicted and actual values ​​of heat dissipation demand under the same operating conditions in the historical fitting period, and calculate the deviation rate between the predicted and actual values ​​of heat dissipation demand for each period; count the deviation rates of all historical fitting periods, calculate the average and maximum values ​​of the deviation rates, and then calculate the upper and lower boundaries of the heat dissipation demand prediction threshold.

5. The cold storage control system based on real-time electricity pricing as described in claim 1, characterized in that, The steps for predicting the range of peak-valley electricity price and cold storage interval for the next cycle include: Based on the period and working day attributes of the next cycle, as well as the characteristics of electricity price changes, historical matching cycles for the next cycle are selected. Retrieve data on peak-valley electricity price and cold storage interval schemes for historical matching cycles, including start time, end time, interval duration, and appropriate cold storage capacity; Clustering operations are performed on the start and end times of the electricity price peak-valley cold storage interval for all historical matching cycles to generate corresponding start time clusters and end time clusters. The data centrality within each cluster is calculated, and the start time cluster and end time cluster with the highest data centrality within each cluster are selected as the core start time cluster and core end time cluster. Based on the time distribution of the core start time cluster and the core end time cluster, the start time range and end time range of the electricity price peak-valley cold storage interval for the next cycle are generated.

6. The cold storage control system based on real-time electricity pricing as described in claim 5, characterized in that, The steps for constructing the temporary cold storage area include: The system continuously collects real-time refrigerant storage capacity and marks the collection timestamp, and combines this with the current natural cooling capacity decay to determine the remaining available refrigerant storage capacity. Calculate the shortest time span from the current moment to the start time of the peak-valley cooling storage interval. Based on the predicted total heat dissipation demand for each time period from the current moment to the start time of the next cycle, sum up the cumulative heat dissipation demand. Compare the remaining available cooling storage capacity with the cumulative heat dissipation demand to calculate the cooling storage gap. If the cold storage shortage is greater than 0, it is determined that there is a cold storage shortage, and the temporary cold storage area construction process is initiated: Based on the cold storage gap, combined with the cold storage efficiency of the refrigerant and the rated cooling power of the refrigeration unit, the temporary cold storage duration required to meet the cold storage gap is calculated. Retrieve electricity price forecast data and forecast fluctuation range from the current time to the predicted start time of the next cycle's cold storage interval, set a temporary off-peak price threshold, and filter out the periods when the lower limit of the predicted electricity price fluctuation range is lower than the temporary off-peak price threshold and the upper limit is not higher than the temporary off-peak price threshold. Combined with the duration of temporary cold storage, define the temporary cold storage interval.

7. The cold storage control system based on real-time electricity pricing as described in claim 6, characterized in that, The step of generating the next cycle's cold storage capacity correction value includes: When the current cycle's peak-valley cold storage interval begins, the initial cold storage capacity carried over from the previous cycle to the current cycle and the actual cold storage consumption of the energy storage battery packs in the current cycle are retrieved. At the same time, if there is a temporary cold storage interval construction operation in the current cycle, the temporary cold storage capacity generated by the temporary cold storage interval is retrieved simultaneously to calculate the actual remaining cold storage capacity at the end of the current cycle. Retrieve the baseline cold storage capacity for the current period and the predicted heat dissipation demand of the energy storage battery pack; and combine this with the total natural cooling capacity decay for the current period to obtain the theoretical remaining cold storage capacity for the current period. By combining the actual remaining cold storage capacity with the theoretical remaining cold storage capacity, the deviation value of the cold storage capacity and the percentage of the cold storage capacity deviation are calculated; and the percentage of the cold storage capacity deviation is converted into a basic correction coefficient. Retrieve the duration of the electricity price peak-valley cold storage interval for the next historical matching cycle and calculate the duration fluctuation range to calibrate the basic correction coefficient and generate the cold storage capacity correction coefficient. The appropriate cooling capacity is calculated by retrieving the predicted total heat dissipation demand of the energy storage battery pack for the next cycle and the predicted threshold for heat dissipation demand. Combined with the cooling capacity correction coefficient, the corrected value for the cooling capacity for the next cycle is generated.

8. A cold storage control method based on real-time electricity pricing, implemented based on the cold storage control system based on real-time electricity pricing as described in any one of claims 1-7, characterized in that, include: The system acquires and periodically stores real-time electricity price data from the power grid, extracts historical periodic electricity price change characteristics, filters historical matching periods, and combines the current period's collected electricity price data change trend with the weighted electricity price data of historical matching periods to generate the current period's predicted electricity price data and the predicted electricity price fluctuation range. The system collects operational data of the energy storage battery pack and, based on historical operational data from the historical fitting period, combines the charging and discharging plan parameters of the pack for the next period to predict the total heat dissipation demand of the energy storage battery pack for the next period and the predicted threshold for heat dissipation demand. Based on the refrigerant storage efficiency and rated cooling power, the appropriate storage capacity and corresponding storage duration are calculated. Combined with the current cycle's predicted electricity price data and the predicted fluctuation range of electricity price, the effective candidate storage range is divided. Based on the benchmark storage duration, the effective candidate storage range is corrected to divide the current cycle's peak-valley storage range. Predict the range of the peak-valley cold storage interval for the next electricity price cycle, and combine it with real-time monitoring of the refrigerant cold storage status to determine whether there is a cold storage gap. If so, construct a temporary cold storage interval. Based on the current cycle's cold storage margin deviation, a correction value for the next cycle's cold storage capacity is generated.

Images